pmc.ncbi.nlm.nih.gov

Some G protein heterotrimers physically dissociate in living cells

Abstract

Heterotrimeric G proteins mediate physiological processes ranging from phototransduction to cell migration. In the accepted model of G protein signaling, Gαβγ heterotrimers physically dissociate after activation, liberating free Gα subunits and Gβγ dimers. This model is supported by evidence obtained in vitro with purified proteins, but its relevance in vivo has been questioned. Here, we show that at least some heterotrimeric G protein isoforms physically dissociate after activation in living cells. Gα subunits extended with a transmembrane (TM) domain and cyan fluorescent protein (CFP) were immobilized in the plasma membrane by biotinylation and cross-linking with avidin. Immobile CFP-TM-Gα greatly decreased the lateral mobility of intracellular Gβ1γ2-YFP, indicating the formation of stable heterotrimers. A GTPase-deficient (constitutively active) mutant of CFP-TM-GαoA lost the ability to restrict Gβ1γ2-YFP mobility, whereas GTPase-deficient mutants of CFP-TM-Gαi3 and CFP-TM-Gαs retained this ability. Activation of cognate G protein-coupled receptors partially relieved the constraint on Gβ1γ2-YFP mobility induced by immobile CFP-TM-GαoA and CFP-TM-Gαi3 but had no effect on the constraint induced by CFP-TM-Gαs. These results demonstrate the physical dissociation of heterotrimers containing GαoA and Gαi3 subunits in living cells, supporting the subunit dissociation model of G protein signaling for these subunits. However, these results are also consistent with the suggestion that G protein heterotrimers (e.g., Gαs) may signal without physically dissociating.

Keywords: cross-linking, fluorescence recovery after photobleaching, G protein-coupled receptors

Heterotrimeric G proteins are known to exist in their inactive state as stable complexes of Gα subunits and Gβγ dimers. Gα subunits cycle between inactive (GDP-bound) and active (GTP-bound) states, and the lifetime of the active state is limited by GTP hydrolysis. Biochemical studies have shown that active G protein heterotrimers dissociate into Gα-GTP and Gβγ subunits in vitro (1). However, it has been argued that G protein subunits may not dissociate under more physiological conditions (2–5), and recent resonance energy transfer (RET) studies have suggested that G protein activation in cells involves subunit rearrangement rather than dissociation (4, 5). Physical dissociation of G protein heterotrimers has not been shown to occur in living cells. To address this question we developed a method to detect protein association and dissociation (Fig. 1A). In this method, one protein of an interacting pair is immobilized in the plasma membrane by an extracellular cross-linking agent. A decrease in the lateral mobility of a second protein [measured by using fluorescence recovery after photobleaching (FRAP)] indicates a binding interaction between the two. The mobility of this second protein is restored only if the partners dissociate. Here, we use this method to show that some G protein subunits physically dissociate in living cells, whereas other heterotrimers appear to remain intact.

Fig. 1.

Fig. 1.

Avidin cross-linking immobilizes CFP-TM-Gα subunits but not Gβ1γ2-YFP dimers when the two are expressed separately. (A) An illustration of the binding partner immobilization assay using FRAP. When protein A (blue) does not interact with protein B (yellow), the lateral mobility of protein B is not changed by the presence of immobile protein A. Protein B fluorescence intensity recovers quickly after photobleaching (red arrow). When the two proteins interact transiently, protein B is partly bound and partly unbound at any time, and FRAP of protein B is slowed. When the two proteins interact stably, protein B is completely bound and thus immobile, as indicated by the lack of fluorescence recovery. (B) An illustration of the predicted topology of CFP-TM-Gα subunits and a confocal image of an avidin cross-linked cell expressing only CFP-TM-GαoA. The expanded region is shown before (pre), immediately after (post), and 2 min (120 sec) after photobleaching of a 3-μm circular ROI; shown by the highlighted circle. Plasma membrane fluorescence in the bleached ROI does not recover, indicating that CFP-TM-GαoA is immobile. (C) Average FRAP curves for CFP-TM-GαoA fluorescence in cells subjected to surface biotinylation (black; n = 13) or biotinylation plus avidin cross-linking (red; n = 11). Photobleaching occurred at time = 5 sec. (D) An illustration of the predicted topology of Gβ1γ2-YFP dimers and a confocal image of an avidin cross-linked cell expressing only these dimers. Plasma membrane fluorescence in the bleached ROI recovers, indicating that Gβ1γ2-YFP dimers remain mobile after avidin cross-linking. (E) Average FRAP curves plotted for Gβ1γ2-YFP fluorescence in cells expressing Gβ1γ2-YFP and subjected to surface biotinylation (black; n = 11) or biotinylation plus avidin cross-linking (red; n = 11). Gray lines in C and E indicate the mean fluorescence intensity ± SEM.

Results and Discussion

G protein heterotrimers and subunits are normally attached to the inner leaflet of the plasma membrane by lipid modifications and are free to diffuse in the plane of this membrane. To make Gα subunits that were susceptible to an extracellular cross-linker, we extended the amino termini of GαoA, Gαi3, and Gαs with a transmembrane (TM) domain, cyan fluorescent protein (CFP), and a cleavable signal sequence. For GαoA and Gαi3, we also incorporated a point mutation to remove the site for pertussis toxin (PTX)-mediated ADP ribosylation so that we could block receptor-mediated activation of native (but not exogenous) Gα subunits by treating cells with PTX (see Materials and Methods). For Gαs, we sometimes used a chimera where the last 5 aa were replaced with the corresponding sequence from Gαq so that we could activate these (Gαsq5) subunits with Gαq-coupled receptors (6), thus bypassing native Gαs. We chose to study these Gα isoforms because several independent studies have variably shown increases or decreases in RET between these subunits and labeled Gβγ subunits (4, 5, 7–10). When expressed in HEK 293 cells, CFP-TM-Gα subunits were localized at the plasma membrane (Fig. 1B). In addition, CFP-TM-Gα subunits were capable of forming functional heterotrimers with Gβγ dimers and interacting productively with G protein-coupled receptors (GPCRs), as indicated by agonist-induced activation of inwardly rectifying potassium (GIRK) channels (CFP-TM-GαoA and CFP-TM-Gαi3) or stimulation of adenylate cyclase (CFP-TM-Gαsq5; Fig. 4 and Supporting Materials and Methods, which are published as supporting information on the PNAS web site). CFP-TM-Gα subunits were immobilized by biotinylating extracellular proteins with a membrane-impermeant, amine-reactive reagent and then exposing cells to soluble avidin. Because avidin is tetravalent and binds biotin (and biotinylated proteins) with high affinity, this procedure cross-links and immobilizes proteins exposed to the extracellular environment. FRAP experiments showed that CFP-TM-Gα subunits were mobile after biotinylation alone but immobile after avidin-mediated cross-linking (Fig. 1 B and C). The calculated diffusion coefficients for CFP-TM-Gα subunits before cross-linking were consistent with lateral diffusion of other transmembrane proteins (e.g., 0.11 ± 0.01 μm2·sec−1 for CFP-TM-GαoA; n = 13) (11).

Gβγ dimers were labeled by using bimolecular fluorescence complementation as described by Berlot and colleagues (12). Gβ1 subunits were fused to a carboxyl-terminal fragment (amino acids 156–239) of a YFP variant, and Gγ2 subunits were fused to an amino-terminal fragment (amino acids 1–155) of YFP. Coexpression of these subunits in HEK cells resulted in the formation of Gβ1γ2 dimers and assembly of intact YFP molecules, as indicated by membrane-localized yellow fluorescence (Fig. 1D). Gβ1γ2-YFP dimers were capable of forming functional heterotrimers with Gα subunits and of interacting with GIRK channels (Fig. 4A). FRAP experiments showed that avidin-mediated cross-linking of cell surface proteins had no effect on the lateral mobility of Gβ1γ2-YFP dimers, as expected for proteins associated with the inner leaflet of the plasma membrane (Fig. 1D and E). The diffusion coefficient for Gβ1γ2-YFP dimers was 0.23 ± 0.03 μm2·sec−1 (n = 11) after biotinylation and 0.23 ± 0.04 μm2·sec−1 (n = 11) after biotinylation and avidin cross-linking (P = 0.95). The diffusion of Gβ1γ2-YFP dimers was faster than the diffusion of CFP-TM-Gα subunits, consistent with the finding that proteins attached to the plasma membrane by lipid modifications diffuse more rapidly than transmembrane proteins (11). This result also shows that Gβ1γ2-YFP does not bind to native proteins that are affected by avidin-mediated cross-linking. Experiments with different-sized photobleached regions (spot size analysis) and fluorescence loss in photobleaching from adjacent regions (data not shown) indicated that fluorescence recovery resulted from lateral movement of G protein subunits rather than insertion of new subunits into the membrane (see also Fig. 3 A and B).

Fig. 3.

Fig. 3.

Receptor activation releases Gβ1γ2-YFP from immobile CFP-TM-GαoA and CFP-TM-Gαi3 but not CFP-TM-Gαsq5. (A) A confocal image of Gβ1γ2-YFP fluorescence in an avidin cross-linked cell expressing A1 adenosine receptors CFP-TM-GαoA and Gβ1γ2-YFP. Cells were pretreated with PTX (300 ng·ml−1) to prevent activation of heterotrimers containing native Gα subunits. The photobleached area is shown as a highlighted circle, and ROIs used for analysis are shown in red (bleached ROI) and blue (adjacent ROI). (Scale bar: 3 μm.) (B) Average fluorescence (a.u.) in the bleached ROI (red) and adjacent ROI (blue) plotted vs. time before and after photobleaching (at time = 5 sec). Application of adenosine (50 μM; horizontal bar) for 30 sec increased fluorescence in the bleached ROI and decreased fluorescence in the adjacent ROI. (C) Summary of experiments (n = 30) with CFP-TM-GαoA identical to that shown in A and B. The mean ratio of Gβ1γ2-YFP fluorescence in the bleached (YFPblch) and adjacent (YFPadj) ROIs (±SEM) is plotted vs. time. (D) Summary of identical experiments (n = 27) with CFP-TM-Gαi3. Adenosine receptor activation induced redistribution of Gβ1γ2-YFP fluorescence from the adjacent region to the bleached region, consistent with physical dissociation from immobile CFP-TM-GαoA and CFP-TM-Gαi3. (E) Summary of experiments (n = 19) in cells expressing V1a vasopressin receptors and CFP-TM-Gαsq5. Arginine vasopressin (1 or 10 μM) was applied for 60 sec where indicated by the horizontal bar. (F) Summary of experiments (n = 17) with constitutively active CFP-TM-Gαsq5 Q227L. Vasopressin receptor activation did not induce redistribution of Gβ1γ2-YFP fluorescence. Note the difference in time scale between A–D and E and F.

To detect the formation of Gαβγ heterotrimers, we expressed Gβ1γ2-YFP dimers together with CFP-TM-Gα subunits. We adjusted the ratio of transfected plasmid DNAs to express an excess of CFP-TM-Gα so that most Gβ1γ2-YFP dimers would bind to CFP-TM-Gα subunits rather than endogenous Gα subunits. Under these conditions, the diffusion coefficient of Gβ1γ2-YFP dimers was reduced to 0.12 ± 0.04 μm2·sec−1 (with CFP-TM-GαoA; n = 8), a value that was not significantly different from that of CFP-TM-GαoA (P = 0.57). Moreover, when expressed with CFP-TM-Gα subunits, the mobility of Gβ1γ2-YFP dimers was dramatically decreased by avidin-mediated cross-linking, consistent with the formation of stable CFP-TM-GαGβ1γ2-YFP heterotrimers (Fig. 2A–D). Control experiments suggested that this effect reflected a specific interaction between inactive CFP-TM-Gα subunits and Gβ1γ2-YFP dimers. Gβ1γ2-YFP dimers remained mobile after cross-linking when coexpressed with Gα subunits lacking a transmembrane extension (Fig. 2H). This finding suggests that heterotrimer formation itself did not render Gβ1γ2-YFP dimers susceptible to immobilization, as might be the case if heterotrimers bound to an immobile membrane protein. In addition, CFP-TM-GαoA subunits bearing mutations shown to prevent Gβγ binding (13, 14), together with a mutation that promotes constitutive activity (15), were unable to significantly restrict the mobility of Gβ1γ2-YFP dimers; normalized fluorescence recovery was 0.80 ± 0.07 (n = 15) for this mutant compared with 0.85 ± 0.08 (n = 8) for biotinylated inactive CFP-TM-GαoA subunits (P = 0.70). This result rules out the possibility that the effect of immobile CFP-TM-Gα on Gβ1γ2-YFP mobility reflected a nonspecific effect of overexpressing and immobilizing a TM-domain protein.

Fig. 2.

Fig. 2.

Association of heterotrimers containing inactive CFP-TM-Gα subunits or GTPase-deficient CFP-TM-Gα subunits and Gβ1γ2-YFP. (A) Average FRAP curves (±SEM) for Gβ1γ2-YFP fluorescence in cells also expressing CFP-TM-GαoA after biotinylation (black; n = 8) or avidin cross-linking (red; n = 11). (B) Average FRAP curves for Gβ1γ2-YFP fluorescence in cells also expressing CFP-TM-Gαi3 after biotinylation (black; n = 10) or avidin cross-linking (red; n = 11). (C) Average FRAP curves for Gβ1γ2-YFP fluorescence in cells also expressing CFP-TM-Gαs after biotinylation (black; n = 12) or avidin cross-linking (red; n = 7). (D) Summary of FRAP expressed as normalized fluorescence recovery 180 sec after photobleaching for the same cells as in A–C. Bars represent the mean (±SEM); ∗, P < 0.01. (E) Average FRAP curves for Gβ1γ2-YFP fluorescence in avidin cross-linked cells also expressing CFP-TM-GαoA (red; n = 16) or constitutively active CFP-TM-GαoA Q205L (black; n = 18). (F) Average FRAP curves for Gβ1γ2-YFP fluorescence in avidin cross-linked cells also expressing CFP-TM-Gαi3 (red; n = 18) or constitutively active CFP-TM-Gαi3 Q204L (black; n = 18). (G) Average FRAP curves for Gβ1γ2-YFP fluorescence in avidin cross-linked cells also expressing CFP-TM-Gαs (red; n = 12) or constitutively active CFP-TM-Gαs Q227L (black; n = 10). (H) Summary of FRAP expressed as normalized fluorescence recovery 180 sec after photobleaching for the same cells as in E–G. “No TM” refers to cells expressing wild-type GαoA (n = 14), Gαi3 (n = 16), or Gαs (n = 10) subunits (without a transmembrane domain). Bars represent the mean (±SEM); ∗, P < 0.01. In all panels, photobleaching occurred at time = 5 sec.

If G protein activation results in physical dissociation of Gα subunits and Gβγ dimers, we predicted that active (GTP-bound) CFP-TM-Gα subunits would lose their ability to restrict the mobility of Gβ1γ2-YFP dimers. To study the interaction between active CFP-TM-Gα subunits and Gβ1γ2-YFP, we introduced a well characterized point mutation (Q to L in switch 3) that dramatically impairs GTPase activity into each Gα subunit. Studies have shown that GTPase-deficient Gα subunits cycle very slowly between inactive and active states, such that a large fraction is in the active state without GPCR activation (16–18). The active mutants of the three Gα isoforms differed markedly in their ability to restrict the lateral mobility of Gβ1γ2-YFP dimers. Gβ1γ2-YFP remained mobile in cross-linked cells expressing constitutively active CFP-TM-GαoA Q205L subunits (Fig. 2 E and H). A trivial explanation for the difference between inactive and active CFP-TM-GαoA subunits is that the latter protein fails to express at levels in the plasma membrane sufficient to decrease Gβ1γ2-YFP mobility. However, fluorescence intensity measurements indicated that the constitutively active mutant expressed slightly better than inactive CFP-TM-GαoA [147 ± 12 vs. 127 ± 10 arbitrary units (a.u.), respectively, n = 20 each; P = 0.21], and Gβ1γ2-YFP dimers expressed equally well in both cases (161 ± 12 vs. 160 ± 11 a.u.; P = 0.97). These results suggest that constitutive activation of GTPase-deficient CFP-TM-GαoA subunits decreases their ability to bind Gβ1γ2-YFP dimers to the point where heterotrimers containing these subunits can fully dissociate. In contrast, Gβ1γ2-YFP mobility was restricted by active CFP-TM-Gαi3 Q204L and CFP-TM-Gαs Q227L subunits to the same degree as by their inactive counterparts (Fig. 2 F–H), suggesting that heterotrimers containing GTPase-deficient CFP-TM-Gαi3 and CFP-TM-Gαs subunits remain intact. This difference could reflect a smaller fraction of mutant CFP-TM-Gαi3 and CFP-TM-Gαs subunits in the active state (e.g., because of slower spontaneous GDP release) in unstimulated cells. Alternatively, this observation is consistent with the idea that subunit dissociation may occur less readily (or not at all) with these Gα isoforms (10).

GTPase-deficient Gα subunits will remain in the active state far longer than Gα subunits with normal GTPase activity, increasing the probability that dissociation will occur before GTP hydrolysis. To determine whether G protein subunits physically dissociate during activation cycles of normal duration, we needed to study Gα subunits with intact GTPase activity. Therefore, we performed experiments to determine whether receptor activation would release Gβ1γ2-YFP dimers from immobile CFP-TM-Gα subunits. Because any inactive CFP-TM-Gα would rapidly rebind free Gβ1γ2-YFP in these experiments, we titrated expression of these subunits (by varying the ratio of transfected plasmid DNAs) to avoid a large excess of the former. To activate CFP-TM-GαoA and CFP-TM-Gαi3, we also expressed A1 adenosine receptors (A1Rs), because these receptors activated heterotrimers containing these subunits (Fig. 4A) and pretreated cells with PTX (300 ng·ml−1) to prevent activation of heterotrimers containing native Gα subunits. To activate CFP-TM-Gαsq5, we also expressed V1a vasopressin receptors, because these receptors activated heterotrimers containing these subunits (Fig. 4B). As was the case in our previous experiments, the mobility of Gβ1γ2-YFP dimers was greatly decreased by avidin cross-linking of all three CFP-TM-Gα isoforms under these conditions. However, agonist activation of coexpressed receptors had isoform-specific effects on Gβ1γ2-YFP mobility. With CFP-TM-GαoA application of the A1R agonist adenosine (50 μM) during FRAP experiments increased Gβ1γ2-YFP mobility, consistent with the physical dissociation of CFP-TM-GαoAGβ1γ2-YFP heterotrimers. This increase in mobility was evident as an adenosine-induced increase in Gβ1γ2-YFP fluorescence recovery in the bleached region of interest (ROI), and an adenosine-induced decrease in fluorescence in adjacent (unbleached) regions of the plasma membrane (Fig. 3B). The absolute bleached/adjacent ROI fluorescence ratio recovered 1.9 ± 0.4% in the 10-sec interval before adenosine application and 6.3 ± 0.7% in the 10-sec interval at the beginning of adenosine application (Fig. 3C; n = 30), indicating an increase in the rate of fluorescence recovery in the bleached ROI. A similar although less robust response was observed with CFP-TM-Gαi3; the ROI ratio recovered 2.4 ± 0.3% during the 10 seconds before adenosine application and 4.7 ± 0.7% during the first 10 seconds of adenosine application (Fig. 3D; n = 27). For both isoforms, the rate of fluorescence recovery during adenosine application appeared to return to the control rate before recovery was complete, suggesting that a fraction of the Gβ1γ2-YFP remained relatively immobile during prolonged agonist application. These results are consistent with the lateral movement of free unbleached Gβ1γ2-YFP dimers into the bleached ROI, which would require that they physically dissociate from immobile CFP-TM-GαoA and CFP-TM-Gαi3. In contrast to these results, application of the V1a vasopressin receptor agonist vasopressin (10 μM) failed to increase Gβ1γ2-YFP mobility when these dimers were immobilized by either inactive CFP-TM-Gαsq5 (Fig. 3E) or GTPase-deficient CFP-TM-Gαsq5 Q227L subunits (Fig. 3F). These results are consistent with the maintained association of Gαs-containing heterotrimers during signaling. In all cases CFP-TM-Gα subunits remained immobile during agonist application, thus agonist-induced redistribution of Gβ1γ2-YFP fluorescence could not have been due to movement of intact heterotrimers. In addition, application of a control (agonist-free) solution had no effect on the rate of fluorescence recovery (data not shown).

Heterotrimeric G protein subunits physically dissociate after activation in vitro (19), and the idea that subunits also dissociate at the plasma membrane is firmly entrenched. However, in vitro experiments are most often carried out with nonhydrolyzable GTP analogs, high concentrations of Mg2+, in the presence of detergent and in the absence of GTPase accelerating proteins (GAPs), i.e., conditions that might artificially favor dissociation. In addition, other biochemical evidence suggests that some activated G proteins can function as intact heterotrimers (2, 3). Therefore, the physical dissociation of G protein subunits in living cells has remained an open question. Renewed interest in this question has been stimulated by recent studies where interactions between G protein subunits have been studied by using RET between labeled subunits. These studies have shown that RET between some Gα subunits (or GPCRs) and Gβγ dimers can increase upon activation (4, 5, 7), suggesting these active subunits rearrange and may not dissociate. In contrast, RET between other Gα subunits and Gβγ dimers decreases upon activation (8–10, 20, 21), an observation consistent with subunit dissociation. However, because energy transfer is sensitive to changes in distance and orientation between molecules, a decrease in RET is equally consistent with subunit rearrangement with maintained association. In addition, because RET reports on (possibly heterogeneous) populations of proteins, a net increase in RET does not rule out dissociation, because the majority of the signal could arise from G protein molecules in an intermediate state (5, 7). For these reasons RET techniques are inherently unable to demonstrate physical dissociation.

We examined this question by using a technique that unambiguously discriminates subunit dissociation from rearrangement, as movement of molecules between bleached and unbleached regions requires complete dissociation from immobile binding partners. We directly demonstrate physical dissociation of active G protein heterotrimers containing CFP-TM-GαoA and CFP-TM-Gαi3 subunits in living cells. A notable difference between these isoforms was that GTPase-deficient CFP-TM-Gαi3 immobilized Gβ1γ2-YFP, whereas GTPase-deficient CFP-TM-GαoA did not. One explanation for this difference is that GαoA subunits have a higher rate of spontaneous GDP release (22), thus a higher proportion of these subunits might have been in the active state. Alternatively, active Gαi3 subunits may be less likely to dissociate from Gβγ subunits than active GαoA subunits. The latter interpretation is consistent with the less pronounced effect of receptor activation observed in this study (Fig. 3D), as well as the different RET changes observed upon agonist activation of these isoforms (10).

Although our results indicate that G protein subunits can physically dissociate after receptor activation in cells, they do not imply that dissociation is required for effector activation or even that dissociation is a general feature of heterotrimeric G proteins in vivo. In fact, our results with Gαs (and Gαsq5) are consistent with the idea that some G protein heterotrimers remain largely intact after activation in cells (4, 5, 10). Receptor activation of even GTPase-deficient CFP-TM-Gαsq5 failed to liberate mobile Gβ1γ2-YFP, suggesting active heterotrimers containing these subunits remained intact. However, this interpretation relies on the unverified assumption that a large fraction of the available CFP-TM-Gαsq5 subunits were driven to the active (GTP-bound) state under these conditions. Additional experiments will be required to determine the extent to which native Gα isoforms physically dissociate under physiological conditions in intact cells.

How might physical dissociation of G protein subunits (or the lack thereof) impact signaling? Because the surfaces involved in interactions between G protein subunits overlap those involved in binding to effectors, it is possible that complete physical dissociation is required for some interactions with effectors. Dissociation might also allow subunits to dissociate from the plasma membrane and translocate to the cytosol (23–25). In addition, G protein signaling is known to be highly divergent, and physical dissociation of G protein subunits would allow signals carried by individual heterotrimers to diverge to downstream targets that are not in close proximity to each other. Finally, the physical exchange of Gβγ dimers between Gα isoforms might also be important for facilitation or inhibition of GPCR signals by activation of other GPCRs, i.e., receptor “cross-talk” (26). On the other hand, signaling by intact heterotrimers could provide a mechanism for Gα subunits to determine receptor-G protein specificity for signals mediated by Gβγ dimers (27). For example, the reluctance of active Gαs to release Gβγ subunits could partly explain how such heterotrimers avoid activating Gβγ effectors normally activated by Gαi- or Gαo-containing heterotrimers. It is also possible that maintained association of G protein heterotrimers is necessary for activation of some effectors, as suggested recently for Gαq and phospholipase Cβ (28). For these reasons, it will be important to determine when physical dissociation of various G protein subunits occurs in living cells. The method introduced here should prove useful for this purpose, as well as for other studies of protein interactions in living cells.

Materials and Methods

cDNA Constructs.

CFP-TM-Gα subunits consisted of (starting at the amino terminus) a cleavable signal peptide from human growth hormone, enhanced CFP, the amino-terminal 103 aa of the rat μ-opioid receptor (including TM 1 and intracellular loop 1) and human Gα subunits. Active CFP-TM-Gα Q to L mutants were identical, with the exception of a single base mutation. CFP-TM-GαoA and CFP-TM-Gαi3 constructs also incorporated C to G mutations at the −4 position to render them insensitive to PTX-mediated ADP-ribosylation. In CFP-TM-Gαsq5 constructs the 5 aa normally found in the carboxyl terminus of Gαs (QYELL) were replaced with those normally found in the carboxyl terminus of Gαq (EYNLV). The Gβγ binding-defective and active mutant of CFP-TM-GαoA was identical to CFP-TM-GαoA with the additional mutations R179C, I19A, E20A and K21A. YN-Gγ2 consisted of amino acids 1–155 of enhanced YFP (EYFP) incorporating the F47L mutation fused to the amino terminus of the bovine Gγ2 gene. YC-Gβ1 consisted of amino acids 156–239 of EYFP fused to amino terminus of bovine Gβ1. Cotransfection of these two constructs drove expression of “Gβ1γ2-YFP.” All constructs were made by using an adaptation of the QuikChange (Stratagene, La Jolla, CA) mutagenesis protocol, were expressed from pcDNA3.1 (Invitrogen, Carlsbad, CA), and were verified by automated sequencing. YN-Gγ2 and YC-Gβ1 were generously provided by Stephen Ikeda and Huanmian Chen (National Institute of Alcohol Abuse and Alcoholism, Rockville, MD). The rat A1R was generously provided by Mark Olah (University of Cincinnati, Cincinnati, OH). The V1a vasopressin receptor and Gα subunits were obtained from the University of Missouri, Rolla, MO) cDNA resource (www.cdna.org).

Cell Culture and Transfection.

HEK 293 cells (American Type Culture Collection, Manassas, VA) were propagated in plastic flasks and on polylysine-coated glass coverslips according to the supplier's protocol. Cells were transfected by using polyethyleneimine and were used for experiments 12–48 h later. For experiments involving receptor activation, pertussis toxin (300 ng·ml−1; List Biological Laboratory, Campbell, CA) was added to the culture medium immediately after transfection.

Avidin-Mediated Cross-Linking.

Cells were rinsed three times in BiotinElation buffer containing 150 mM NaCl, 2.5 mM KCl, 10 mM Hepes, 12 mM glucose, 0.5 mM CaCl2, and 0.5 mM MgCl2 (BE buffer; pH 8.0) and incubated at room temperature for 15 min in BE containing 0.5 mg·ml−1 NHS-sulfo-LC-LC-biotin (Pierce, Rockford, IL). Cells were washed an additional three times and incubated for 15 min in 0.1 mg·ml−−1 avidin (Pierce). Control cells were biotinylated but were not exposed to avidin. Avidin cross-linked cells remained viable and endocytosed avidin after several hours, thus all experiments were performed within 1 h of avidin exposure.

Imaging.

Coverslips bearing control or avidin cross-linked cells were transferred to the stage of a SP2 scanning confocal microscope (Leica, Wetzlar, Germany) and imaged by using a ×63, 1.4 NA objective. Cells were excited by using 458 nm (for CFP) or 512 nm (for YFP) laser lines set to underfill the back aperture of the objective, and the confocal pinhole was opened to ≈3 airy units. Low-intensity illumination was used during a control (prebleach) period, after which a 3-μm segment of the plasma membrane edge was irreversibly photobleached by increasing the laser intensity. The illumination intensity for photobleaching was adjusted to produce ≈50% loss of fluorescence in the bleached region. Recovery of fluorescence into the bleached segment of plasma membrane was monitored for 3–5 min by using low-intensity illumination. Average pixel intensity in the bleached region was corrected for photobleaching during low-intensity illumination, normalized, and plotted vs. time. Diffusion coefficients (D) were derived by fitting fluorescence recovery curves to the empirical equation: F(t) = Fi + Fm (1 − [w2(w2 + 4πDt)−1]0.5), where Fi is the fluorescence intensity immediately after photobleaching, Fm is the intensity recovered, w is the width of the bleached plasma membrane, and D is the effective one-dimensional diffusion coefficient (29). For experiments involving receptor activation, cells were continuously perfused with a solution containing 150 mM NaCl, 5 mM KCl, 10 mM Hepes, 10 mM glucose, 1.5 mM CaCl2, and 2.5 mM MgCl2 (pH 7.2). Solution changes were made by using a multiport attachment and perfusion capillary positioned directly in front of the cell under study.

Statistical Analysis.

Comparison of normalized fluorescence recovery (Fig. 2) was carried out by using one-way ANOVA, followed by Bonferroni comparisons; statistical significance (indicated by an asterisk) was defined as P < 0.01. Other comparisons were made by using Student's t test.

Supplementary Material

Supporting Information

Acknowledgments

We thank Stephen Ikeda and Huanmian Chen for generously providing plasmids encoding YN-Gγ2 and YC-Gβ1, Mark Olah for generously providing the plasmid encoding the rat A1R, L. Y. Jan (University of California, San Francisco, CA) for generously providing HEK cells stably expressing GIRK1 and GIRK4, and J. Dempster (University of Strathclyde, Glasgow, U.K.) for WinWCP electrophysiology software. This work was supported by Ruth L. Kirschstein National Research Service Award NS052070 (to R.M.L), American Heart Association Predoctoral Fellowship 0515176B (to R.M.L), National Institutes of Health Grant NS36455 (to N.A.L.), and National Science Foundation Grant MCB-0620024 (to N.A.L.).

Abbreviations

GPCR

G protein-coupled receptor

GIRK

G protein-activated inwardly rectifying potassium channel

FRAP

fluorescence recovery after photobleaching

CFP

cyan fluorescent protein

PTX

pertussis toxin

RET

resonance energy transfer

TM

transmembrane.

Footnotes

The authors declare no conflict of interest.

This article is a PNAS direct submission.

References

  • 1.Gilman AG. Annu Rev Biochem. 1987;56:615–649. doi: 10.1146/annurev.bi.56.070187.003151. [DOI] [PubMed] [Google Scholar]
  • 2.Levitzki A, Klein S. ChemBioChem. 2002;3:815–818. doi: 10.1002/1439-7633(20020902)3:9<815::AID-CBIC815>3.0.CO;2-E. [DOI] [PubMed] [Google Scholar]
  • 3.Rebois RV, Warner DR, Basi NS. Cell Signal. 1997;9:141–151. doi: 10.1016/s0898-6568(96)00133-7. [DOI] [PubMed] [Google Scholar]
  • 4.Bunemann M, Frank M, Lohse MJ. Proc Natl Acad Sci USA. 2003;100:16077–16082. doi: 10.1073/pnas.2536719100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gales C, Rebois RV, Hogue M, Trieu P, Breit A, Hebert TE, Bouvier M. Nat Methods. 2005;2:177–184. doi: 10.1038/nmeth743. [DOI] [PubMed] [Google Scholar]
  • 6.Conklin BR, Herzmark P, Ishida S, Voyno-Yasenetskaya TA, Sun Y, Farfel Z, Bourne HR. Mol Pharmacol. 1996;50:885–890. [PubMed] [Google Scholar]
  • 7.Gales C, Van Durm JJ, Schaak S, Pontier S, Percherancier Y, Audet M, Paris H, Bouvier M. Nat Struct Mol Biol. 2006;13:778–786. doi: 10.1038/nsmb1134. [DOI] [PubMed] [Google Scholar]
  • 8.Gibson SK, Gilman AG. Proc Natl Acad Sci USA. 2006;103:212–217. doi: 10.1073/pnas.0509763102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Azpiazu I, Gautam N. J Biol Chem. 2004;279:27709–27718. doi: 10.1074/jbc.M403712200. [DOI] [PubMed] [Google Scholar]
  • 10.Frank M, Thumer L, Lohse MJ, Bunemann M. J Biol Chem. 2005;280:24584–24590. doi: 10.1074/jbc.M414630200. [DOI] [PubMed] [Google Scholar]
  • 11.Kenworthy AK, Nichols BJ, Remmert CL, Hendrix GM, Kumar M, Zimmerberg J, Lippincott-Schwartz J. J Cell Biol. 2004;165:735–746. doi: 10.1083/jcb.200312170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Hynes TR, Tang L, Mervine SM, Sabo JL, Yost EA, Devreotes PN, Berlot CH. J Biol Chem. 2004;279:30279–30286. doi: 10.1074/jbc.M401432200. [DOI] [PubMed] [Google Scholar]
  • 13.Denker BM, Neer EJ, Schmidt CJ. J Biol Chem. 1992;267:6272–6277. [PubMed] [Google Scholar]
  • 14.Slepak VZ, Wilkie TM, Simon MI. J Biol Chem. 1993;268:1414–1423. [PubMed] [Google Scholar]
  • 15.Wong YH, Federman A, Pace AM, Zachary I, Evans T, Pouyssegur J, Bourne HR. Nature. 1991;351:63–65. doi: 10.1038/351063a0. [DOI] [PubMed] [Google Scholar]
  • 16.Masters SB, Miller RT, Chi MH, Chang FH, Beiderman B, Lopez NG, Bourne HR. J Biol Chem. 1989;264:15467–15474. [PubMed] [Google Scholar]
  • 17.Graziano MP, Gilman AG. J Biol Chem. 1989;264:15475–15482. [PubMed] [Google Scholar]
  • 18.Kroll SD, Chen J, De Vivo M, Carty DJ, Buku A, Premont RT, Iyengar R. J Biol Chem. 1992;267:23183–23188. [PubMed] [Google Scholar]
  • 19.Northup JK, Smigel MD, Sternweis PC, Gilman AG. J Biol Chem. 1983;258:11369–11376. [PubMed] [Google Scholar]
  • 20.Janetopoulos C, Jin T, Devreotes P. Science. 2001;291:2408–2411. doi: 10.1126/science.1055835. [DOI] [PubMed] [Google Scholar]
  • 21.Yi TM, Kitano H, Simon MI. Proc Natl Acad Sci USA. 2003;100:10764–10769. doi: 10.1073/pnas.1834247100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Linder ME, Ewald DA, Miller RJ, Gilman AG. J Biol Chem. 1990;265:8243–8251. [PubMed] [Google Scholar]
  • 23.Iiri T, Backlund PS, Jr, Jones TL, Wedegaertner PB, Bourne HR. Proc Natl Acad Sci USA. 1996;93:14592–14597. doi: 10.1073/pnas.93.25.14592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ransnas LA, Svoboda P, Jasper JR, Insel PA. Proc Natl Acad Sci USA. 1989;86:7900–7903. doi: 10.1073/pnas.86.20.7900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Akgoz M, Kalyanaraman V, Gautam N. J Biol Chem. 2004;279:51541–51544. doi: 10.1074/jbc.M410639200. [DOI] [PubMed] [Google Scholar]
  • 26.Quitterer U, Lohse MJ. Proc Natl Acad Sci USA. 1999;96:10626–10631. doi: 10.1073/pnas.96.19.10626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Leaney JL, Milligan G, Tinker A. J Biol Chem. 2000;275:921–929. doi: 10.1074/jbc.275.2.921. [DOI] [PubMed] [Google Scholar]
  • 28.Evanko DS, Thiyagarajan MM, Takida S, Wedegaertner PB. Cell Signal. 2005;17:1218–1228. doi: 10.1016/j.cellsig.2004.12.008. [DOI] [PubMed] [Google Scholar]
  • 29.Ellenberg J, Siggia ED, Moreira JE, Smith CL, Presley JF, Worman HJ, Lippincott-Schwartz J. J Cell Biol. 1997;138:1193–1206. doi: 10.1083/jcb.138.6.1193. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information